Population profile and residential environment of an urban poor community in Dhaka, Bangladesh
Khalequzzaman et al. Environmental Health and
Population profile and residential environment of an urban poor community in Dhaka, Bangladesh
Chifa Chiang 0
Bilqis Amin Hoque
Sohel Reza Choudhury
Hiroshi Yatsuya 0
Yoshihisa Hirakawa 0
Syed Shariful Islam
Atsuko Aoyama 0
0 Department of Public Health and Health Systems, Nagoya University School of Medicine , Nagoya , Japan
Objectives: A population survey was conducted in an urban shantytown in Bangladesh, as a baseline study of future epidemiological studies. This paper aims to describe the findings of the study, including the population profile and residential environment of the urban poor. Methods: We conducted a complete count household survey in an urban poor community in Dhaka. Using a brief structured questionnaire in Bengali language, trained interviewers visited each household and asked questions such as: duration of residence; ownership of house, toilet and kitchen; water supply; number of family members; age, sex, education, occupation, tobacco use, and history of diseases of each family member. Results: We found that there were 8604 households and 34,170 people in the community. Average number of household members was 4.0. Most people had access to safe water, but only 16% lived in the house with a toilet. Based on the proxy indicators of household wealth levels, we identified that about 39% were relatively well-off, while the rest were very poor. Tobacco use was prevalent in men regardless of age and in women aged over 35 years. Prevalence of self-reported hypertension and diabetes was slightly higher in women than in men, although over 70% of the respondents didn't know if they had such diseases. Incidences of diarrhea in the last one month were relatively low. Conclusions: The study showed population profile and sanitation environment in an urban poor community by a complete count survey. We expect the study to serve as a baseline for future epidemiological studies.
The urban poor; Bangladesh; Baseline population survey; Residential environment; Non-communicable diseases
Environmental Health and
Bangladesh is a lower-middle income country in South
Asia, with over 160 million population in 2015 . The
population of the metropolitan area of the capital city
Dhaka was around 17 million. Urban population in
Bangladesh is rapidly increasing, as indicated by 3.4%
annual urban population growth in comparison with
1.2% population growth in the whole nation in 2015.
Infectious diseases are still prevalent in Bangladesh,
mostly due to poor sanitation environment. In addition,
the burden of non-communicable diseases (NCDs) is
also increasing: age-standardized mortality rates of all
NCDs, cardiovascular diseases, and diabetes were 548.9,
166.2, and 29.8 per 100,000 population in 2012,
respectively . Previous surveys on NCD risk factors in
Bangladesh showed that the prevalence was higher in
urban areas than in rural areas [3–7], and NCDs were
prevalent even among the poor in a rural area .
However, little has been known about the situation of
NCD risk factors among the poor living in urban
shantytowns, where lifestyles are changing rapidly. In addition,
low birth weight and childhood malnutrition might be
more prevalent among the poor, which may reportedly
increase the risks of cardiovascular diseases and diabetes
in adulthood [9, 10]. Therefore, the urban poor could be
considered as high risk population, but proper NCD
control measures are difficult to be taken. One of the
© The Author(s). 2017 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0
International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and
reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to
the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver
(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
reasons of this difficulty is the unknown situation of the
Since no reliable recent census are available in
shantytowns where people move in and out without registration,
it is difficult to conduct epidemiological surveys based on
representative sampling methods. Therefore, we conducted
a population survey targeting all households in a
shantytown as a baseline of future epidemiological surveys. In
addition to demographic data, we also collected some
information related to residential environment and awareness of
NCDs, as they may help us make a plan of future studies.
This paper aims to describe the findings of the
baseline study, including population profile, residential
sanitation environment, and prevalence of self-reported
NCD risk factors among the urban poor.
Study site and study population
We conducted a census-like baseline population study in
Bauniabadh, an urban poor community in Mirpur ward,
Dhaka city, Bangladesh [11–13]. The community was
originally established by the government in 1972 as a
settlement for the poor. A same size land plot (about
8.9 m ) was allocated to each household at an affordable
price. Since then, many residents moved in or out without
registration, and the community expanded irregularly with
sprawling shantytowns in and outside the original
boundary. Before this study, it was estimated that there were
about 15,000 households (about 50,000 population) within
the original boundary and about 5000 households in the
areas adjacent to the original boundary.
We targeted all households within the original
boundary but excluded the surrounding shantytowns,
considering the feasibility of the study (Fig. 1). The original area
was consisted of 5 blocks; each block consisted of 22
lanes (except a block consisted of 24 lanes); each lane
consisted of 24 plots. Although each household allocated
each one plot originally, some plots were combined to
build multistoried buildings and it was not known how
many households shared the building.
Staff training and community mobilization
We recruited 14 field interviewers, who lived in the same
community and completed secondary level education, and
Fig. 1 Map of the target community. The framed area is the target community, Bauniabadh. (Source: Urban Planning Department, Dhaka North
City Corporation, Dhaka, Bangladesh)
trained them for interview skills. Two field supervisors
were assigned for managing field activities and controlling
Mobilizing the community and encouraging people to
participate in the survey, meetings with community
leaders and other representatives were held in the
community several times before and during the survey
period. Community leaders were actively involved in
motivating people to participate.
We conducted the baseline population survey from
August to December, 2014. We targeted all households
living in the original 5 blocks of Bauniabadh, including
tenants in rented and sublet houses, and people living
apart from their original households.
We prepared a brief structured questionnaire in
Bengali language. Each household was asked following
questions: type of housing structure, owner or tenant,
duration of residence, ownership of toilet and kitchen,
source of water supply, satisfaction with water quality, and
number of household members. Questions about each
household member were as follows: age, relation to the
household head, sex, religion, education, occupation,
marital status, tobacco use, and history of diseases
(hypertension, diabetes, heart diseases, stroke, and diarrhea in
the last 1 month). Unknown categories for the history
indicated that the respondent did not know if they had
hypertension, diabetes, heart disease, or stroke. The
questionnaire was revised several times until all the
interviewers were confirmed to be able to confidently complete
The trained interviewers visited each household of the
community and conducted face to face interviews in
Bengali language to the household head, or a household
member who could give information of all household
members if the household head was unavailable. The
interviewer also observed housing structure, kitchen,
toilet, and water supply of the household. The supervisors
monitored the interviewers intermittently in the field to
make sure the questions were properly administered.
To avoid overlapping or skipping of households in the
crowded shantytown, we used the following procedures.
We first selected one lane of each block. The interview
started from the first household of the north-west corner
of the selected lane. Before starting the interview, the
interviewer got permission of the resident and marked
an identification code number on the edge of the main
entrance door with a permanent marker, and noted the
same code number in the questionnaire. After finishing
the interview, the interviewer selected the next
household on the right side. Once the interviews of all
households in the right side of the lane were completed, the
interviewer turned around and started the interviews of
the households on the other side of the lane. After
finishing the lane, the interviewer reported to the
supervisor, and the supervisor confirmed all households in the
lane had been covered. It took at least 4 days to
complete interviews of all households in one lane.
In case of multistoried building, the interview started
the household on the right side of the stair at the ground
floor. The interviewer moved to upwards visiting
household at each floor and reached the household at the top
floor. Then the interviewer turned around and
interviewed households on the other side of the stair from
the top floor to the ground floor.
If a sublet family lived together in the same house, the
household interview was conducted to the sublet family
separate from the host family. In case the house was
locked or no one was available for the interview, the
interviewer marked the code number on the door and
noted the number on the questionnaire. The interviewer
returned the household when the household member
was available, and conducted the interview using the
The household residents’ names are separated from the
original sheets, which are coded with serial numbers.
The anonymized data were inputted into a programmed
data entry template and subjected to statistical analyses.
All of the statistical analyses were performed using the
statistical software, IBM SPSS Statistics for Windows,
Version 23.0 (IBM Corp, Armonk, NY, USA).
Differences between all men and all women were tested by
using age-adjusted binary and multinominal logistic
regression models for dichotomous and trichotomous
This study was reviewed and approved by the Bioethics
Review Committee of Nagoya University School of
Medicine, Japan (approval no. 2014-0021). Institutional
Review Boards of Bangabandhu Sheikh Mujib Medical
University and National Heart Foundation Hospital and
Research Institute, Bangladesh, approved the study as
well. Written informed consents were obtained from all
participants after adequate explanations of the study.
We found that there were 8604 households in the 5
blocks of the original Bauniabadh and 34,170 people
(men 17,041; women 17,129) resided. The household
number and population were much less than we
estimated before the study.
Table 1 shows demographic and residential
characteristics of the study population (n = 34,093 excluding 77
without age information). Average number of household
Table 1 Demographic and residential characteristics of the
Mean number of household members
Dwelling since 1972 (%)
New comers in adults aged ≥18 yearsa (%)
Years of formal education in adults aged ≥18 years (%)
Occupation in adults aged ≥18 years (%)
Type of housing structure (%)
Concrete roofs/brick walls/concrete floors (pucca)
Tin roofs/brick walls/mud or wooden floors (semi-pucca)
House ownership (%)
Self-owned toilet (%)
Self-owned kitchen (%)
Source of drinking water (%)
Shallow tube well
Satisfaction with water quality (%)
aThose who had lived in the community for ≤5 years
members was 4.0, suggesting that most households in
the urban poor community were nuclear families and
average fertility was similar to the national total fertility
rate (2.2 in 2013) . About 25% of adults lived in the
community for 5 years or less, indicating frequent
migration of the urban poor. Only 0.4% lived there since
the community was established, and 35% were born in
Figure 2 shows population by sex and age groups,
forming a constrictive type of population pyramid. While
the pyramid indicated that birth rates were decreasing,
the population was still young, as about 80% were less
than 40 years of age.
Figure 3 shows examples of types of housing structure.
About 39% lived in single or multistoried houses with
concrete roofs, concrete floors, and brick walls (pucca),
30% in houses with tin roofs, mud or wooden floors, and
brick walls (semi-pucca), and 31% in houses with tin
roofs, mud or wooden floors, and walls made of thatch
or bamboo (kutcha) (Table 1). It was found that about
84% of the population shared toilets, and 81% shared
kitchens with other households. Almost all the
population used piped water as their source of drinking water,
and 62% satisfied with the water quality.
To categorize household wealth levels, we referred to
the 2010 STEPS survey’s wealth index constructed from
the asset information including the type of main material
used for the roof, wall and floor of the main house and
household ownership of electricity, flush toilet,
television, refrigerator, car, etc. . Adapting the national
survey’s asset information to the urban poor community
where diversity in wealth might be less than that of the
target of the nation-wide survey, we used the type of
housing structure as a proxy indicator of the household
wealth, and categorized household wealth levels into 2
groups: “lower-middle wealth” households were defined
as those living in pucca; and “low wealth” households
were defined as those living in semi-pucca or kutcha.
Lower-middle wealth households tended to have their
own kitchens and toilets, while several low wealth
households shared a kitchen and a toilet. We found that
39% of the population of the community were the
lower-middle wealth group, while 61% were the low
About 52% of adults had at least 5 year education, and
28% had secondary or higher level education. About 26%
of adults were employed, 19% were day laborers, and
16% were self-employed.
Table 2 shows self-reported health related indicators of
adults aged 18 years and older who responded the
interview representing each household (n = 7616), by age
group and sex. The number of women was much larger
than that of men among the respondents, as women
were more likely to be at home when interviewers
visited, while men tended to be working outside. Tobacco
product use was prevalent in men regardless of age and in
women aged over 35 years, although they might be
underreported. Cigarette smoking was prevalent in men
regardless of age, but very few women smoked. Smokeless
tobacco chewing was prevalent both in men and women
aged over 35 years.
Prevalence of self-reported hypertension and diabetes
were 5.9 and 3.6% in men and 8.7 and 4.6% in women,
respectively, both higher in women than in men.
Prevalence of self-reported heart disease and stroke were 2.6
and 4.0% in men and 4.1 and 3.7% in women,
respectively. Around 70 to 90% of the population responded
Fig. 2 Population by sex and age groups
that they did not know if they had hypertension,
diabetes, heart disease, or stroke. A few people in all age
groups reported that they had stroke, however, it was
likely that they confused all types of fainting incidents as
stroke. Incidences of diarrhea in the last one month
were about 3% in men and around 5% in women.
This study showed population profile and residential
environment of the people living in an urban poor
community in Bangladesh. Several demographic
registration systems were established in some rural areas in
Bangladesh , however, complete count surveys in very
Fig. 3 Types of housing structure
Smoking cigarettes (%)
Chewing smokeless tobacco (%)
Tobacco product use (%)
Heart Disease (%)
aDifferences between all men and all women were tested by using age-adjusted binary and multinominal logistic regression models for dichotomous and trichotomous
bSuffering from diarrhea in the past 30 days
crowded urban shantytowns were rare. The recent 2011
census, which took several years to complete, might not
reflect the profile of often fluctuating urban poor
population in details .
The population pyramid was partly similar to the
constricted and stationary types, but had a large proportion
of younger population and a small proportion of elder
population. The population composition was different
from the expansive type reflecting high fertility and high
mortality, or the stationary and constrictive types in high
income countries where both low fertility and low
mortality were achieved. The population pyramid might
reflect declined fertility in the community, while
mortality of elder people had not much decreased. The
population composition might result from the influx of
young population to the urban community from other
areas. Population changes caused by migration might be
more significant than the natural changes in the urban
shantytown [15, 16].
About half of the adults did not complete primary level
education and 32% had never received formal education,
which was consistent with the adult literacy rate (61% in
2013) reported by the Bangladesh government .
However, primary school enrollment of the children in the
community is expected to be high, as there are primary
schools in the community, in line with the national
average (98% primary school enrollment in 2014).
Ratio of employed people was higher than day laborers
or self-employed, perhaps because the community
located close to garment factories, which hired many
manual laborers . Most day laborers involved in hard
physical works such as puling cycle rickshaws, and
selfemployed included street vendors of snacks, tobacco,
and daily commodities. Married women tended not to
work outside as shown that 28% of adults were
We found that households in the urban poor
community were not equally poor anymore, although their
wealth levels were similar when the community was
established 44 years ago. Along with the overall
economic development, some of the current residents were
relatively well-off by buying up several plots to build
brick houses, while others remain very poor sharing
shanties made of bamboo and tin. It was difficult to
estimate incomes of the people in shantytowns, of whom
majority were day laborers, however, we could categorize
household wealth levels by using types of housing as a
proxy indicator. We found that about 39% of the
population was relatively well-off, although they still lived in
the shanty area, perhaps because it was convenient to
stay in the middle of the city.
Although 84% of the population did not have toilets in
their house, most people had access to safe piped water.
Each household was provided a tube well by the
government when the community established 44 years ago.
Then, several environmental improvement programs
sponsored by aid agencies had been conducted in the
area [11, 12], which included providing piped water and
cleaning sewage ditches. Although arsenic
contamination of groundwater was widely observed in Bangladesh
, piped water supply in the community had been
confirmed to be free of arsenic contamination. Good
access to safe water was supposed to bring the relatively
low incidence of diarrhea.
A previous study on urban slums in 6 cities in
Bangladesh reported household conditions in Dhaka
slums as follows: about 92% used piped water supply,
99% shared toilet, only 12% owned house, 46% lived in
kutcha and 52% lived in semi-pucca . Our findings
showed that living conditions of our target area were
much better than those of 4966 slum clusters studied in
2005. This may be attributable to significant
improvement of living conditions in Dhaka in the past decade
along with rapid economic development. This may also
indicate the relatively good living condition of our
targeted area, which was originally a settlement established
by the government.
Another study on health and living conditions in 8
Indian cities reported that household conditions in
Kolkata slums as follows: 85% used piped water supply,
75% shared toilet, and only 7% lived in kutcha or
semipucca . The study also reported that not all slum
dwellers were poor, and proportion of the population of
under 15 years of age was less than 30% as fertility
declined, while the proportion of 15 to 59 age group was
high due to migration. These findings in India were
similar to our findings.
High prevalence of tobacco use was confirmed in this
study, in consistent with previous studies in Bangladesh
[21, 22]. Women chewed tobacco more often than men,
but refrained smoking cigarettes. This indicated that
chewing tobacco products were culturally tolerated,
considering that Bangladesh women observed various
cultural norms and behavioral restrictions. Tobacco
chewing seemed to be more commonly practiced among
people aged over 35 years than young people in the 20s,
suggesting that tobacco chewing was a traditional old
habit while cigarette smoking was a relatively new habit.
Different approaches for men and women, and for young
and old, need to be developed to control tobacco.
The observed low prevalence of hypertension
compared to the crude estimate by WHO (men 22.0%;
women 21.0%)  may be resulted from suboptimal
awareness of the disease, since majority of respondents
in the present study did not know whether they had
hypertension or not. Similarly, the prevalence of diabetes
was about half of the WHO estimated national
prevalence (men 8.6%; women 7.4%) . Furthermore, the
accuracy of the self-report is unknown.
The high ratio of the people who did not know they
had NCDs or not may be in part due to the health
systems unprepared to work on NCD control and
management. Health systems had been designed to
target infectious diseases and maternal and child health,
thus health professionals in the community often lacked
proper trainings for managing NCDs. Most of the clients
of community health centers were women and children,
therefore, adult men tended to use casual health
checkups provided by auxiliary pharmacists or street vendors
, whose diagnosis of hypertension and diabetes were
unlikely to be reliable. Higher prevalence or awareness
in women than in men may suggest more frequent use
of health services by women than men, e.g., blood
pressure measurements during antenatal care. Further
studies are required to investigate prevalence of NCD
The high prevalence of unawareness may be also due
to the lack of opportunities for the people to obtain
accurate knowledge of NCDs. Since most NCDs do not
show specific symptoms in the beginning, it is very
important to provide health education about risk factors
and early symptoms of NCDs. The prevalence of
unawareness was higher in the young than the old, despite
their higher school attendance. Health education for
NCD prevention should be incorporated into formal
The strength of this study was that we conducted
complete count survey targeting an urban shantytown,
where households were too crowded to survey accurately
and population often fluctuated due to migration. We
expect the findings of this study would be utilized as a
baseline population profile for the future epidemiological
studies in the area. We have already conducted an
epidemiological study of NCD risk factors using
representative sampling methods based on the baseline data.
The limitation of this study was that we targeted only
one urban poor community, excluding surrounding
shantytowns, thus the situation might be somewhat
different from that of other urban poor communities.
Another limitation was that we interviewed only one
person in the household, who might have given incorrect
information of other household members. The
respondent in the household was not chosen through a
representative sampling methods, either. Furthermore, all
health related information were self-reported based on
the past history, but were not clinically confirmed,
therefore, current conditions might be underreported.
In conclusion, this study showed population profile
and residential environment in an urban poor
community by a complete count survey. We expect the study to
serve as a baseline for future epidemiological studies.
The authors wish to thank people in the Bauniabadh community, staff
members of the field survey and Environmental and Population Research
Center, for assistance in data collection and valuable advice during the survey.
This work was supported by Grant-in-Aid for Scientific Research (KAKENHI)
[A, 25257505 to AA] from Japan Society for the Promotion of Science.
SRC, AA, MK, BAH, and AM designed the study, MK, SN, BAH, SRC and SSI
conducted the field survey and data collection, CC, MK, HY, SRC, and AA
statistically analyzed and interpreted the data, AA, MK, and CC drafted the
manuscript, HY, SRC, BAH, SSI, HI, AM, and YH provided critical inputs on the
draft. All authors read and approved the final manuscript.
The authors declare that they have no competing interests.
Ethics approval and consent to participate
All procedures performed in studies involving human participants were in
accordance with the ethical standards of the institutional research committees.
Springer Nature remains neutral with regard to jurisdictional claims in
published maps and institutional affiliations.
1. The World Bank. World Development Indicators , Bangladesh. http://data. worldbank.org/country/bangladesh. Accessed 17 Aug 2016 .
2. World Health Organization (WHO). Global Health Observatory data repository, Noncommunicable diseases, Mortality . http://apps.who.int/gho/ data/node.main. A859?lang=en . Accessed 17 Aug 2016 .
3. Rahman M , Flora MS , Akter SFU , Hossain S. Mascie-Taylor CGN . Behavioral risk factors of non-communicable diseases in Bangladesh - tobacco usage, dietary pattern and physical activity status . http://www.who.int/chp/steps/ BangladeshSTEPSReport.pdf?ua=1. Accessed 21 Nov 2016 .
4. Bangladesh Society of Medicine, WHO Country Office for Bangladesh, Ministry of Health and Family Welfare. Non-communicable disease risk factor survey , Bangladesh 2010 . New Delhi: WHO Regional Office for South-East Asia ; 2011 .
5. Zaman MM , Bhuiyan MR , Karim MN , Zaman M , Rahman MM , Akanda AW , et al. Clustering of non-communicable diseases risk factors in Bangladeshi adults: an analysis of STEPS survey 2013 . BMC Public Health . 2015 ; 15 : 659 .
6. Zaman MM , Choudhury SR , Ahmed J , Talukder MH , Rahman AHMS . Blood glucose and cholesterol levels in adult population of Bangladesh: results from STEPS 2006 survey . Indian Heart J . 2016 ; 68 : 52 - 6 .
7. Zaman MM , Rahman M , Rahman M , Bhuiyan MR , Karim M , Chowdhury MJ . Prevalence of risk factors for non-communicable diseases in Bangladesh: results from STEPS survey 2010 . Indian J Public Health . 2016 ; 60 : 17 - 25 .
8. Khan JAM , Trujillo AJ , Ahmed S , Siddiquee AT , Alam N , Mirelman AJ , et al. Distribution of chronic disease mortality and deterioration in household socioeconomic status in rural Bangladesh: an analysis over a 24-year period . Int J Epidemiol . 2015 ; 44 : 1917 - 26 .
9. Risnes KR , Vatten LJ , Baker JL , Jameson K , Sovio U , Kajantie E , et al. Birthweight and mortality in adulthood: a systematic review and meta-analysis . Int J Epidemiol . 2011 ; 40 : 647 - 61 .
10. Whincup PH , Kaye SJ , Owen CG , Huxley R , Cook DG , Anazawa S , et al. Birth weight and risk of type 2 diabetes: a systematic review . JAMA . 2008 ; 300 : 2886 - 97 .
11. United Nations University Institute for the Advanced Study of Sustainability (UNU-IAS), Environment and Population Research Centre . Water and sanitation in an urban poor settlement: a case study of Bauniabad , Bangladesh. Tokyo: UNU-IAS ; 2005 .
12. Sato-Yamamoto Y , Suzuki K , Hoque BA . Sanitation in urban poor settlement and the importance of education for the reduction of the diffused pollution - a case study of Bauniabad, Bangladesh . J Water Environ Technology . 2005 ; 3 : 199 - 207 .
13. Ahmed T , Mahfuz M , Islam MM , Mondal D , Hossain MI , Ahmed AMS , et al. The MAL-ED cohort study in Mirpur, Bangladesh. Clin Infect Dis . 2014 ; 59 : S280 - 6 .
14. Bangladesh Bureau of Statistics. Population and housing census 2011 - community report: Dhaka . Dhaka: Statistics and Informatics Division , Ministry of Planning, Government of Bangladesh; 2015 .
15. Koehlmoos TP , Uddin MJ , Ashraf A , Rashid M. Homeless in Dhaka: violence, sexual harassment, and drug-abuse . J Health Popul Nutr . 2009 ; 27 : 452 - 61 .
16. Lerch M. The role of migration in the urban transition: a demonstration from Albania . Demography. 2014 ; 51 : 1527 - 50 .
17. General Economics Division. Millennium development goals Bangladesh progress report 2015 . Dhaka: Bangladesh Planning Commission, Government of Bangladesh ; 2015 .
18. Smith AH , Lingas EO , Rahman M. Contamination of drinking-water by arsenic in Bangladesh: a public health emergency . Bull World Health Organ . 2000 ; 78 : 1093 - 103 .
19. Centre for Urban Studies (CUS), National Institute of Population Research and Training , MEASURE Evaluation. Slums of urban Bangladesh: mapping and census , 2005 . Dhaka: CUS; 2006 .
20. Gupta K , Arnold F , Lhungdim H. Health and living conditions in eight Indian cities: National Family Health Survey (NFHS-3) India 2005-06 . Mumbai: International Institute for Population Sciences; 2009 .
21. Nargis N , Thompson ME , Fong GT , Driezen P , Hussain AKMG , Ruthbah UH , et al. Prevalence and patterns of tobacco use in Bangladesh from 2009 to 2012: evidence from International Tobacco Control (ITC) study . PLoS One . 2015 ; 10 :e0141135.
22. Pallipudi KM , Sinha DN , Choudhury S , Zaman MM , Asma S , Andes L , et al. Predictors of tobacco smoking and smokeless tobacco use among adults in Bangladesh . Indian J Cancer . 2012 ; 49 : 387 - 92 .
23. World Health Organization . WHO Diabetes country profiles 2016 , Bangladesh. http://www.who. int/diabetes/country-profiles/bgd_en.pdf?ua=1. Accessed 17 Aug 2016 .
24. Aoyama A , Matsuyama A , Al-Shoaibi AAA , Chiang C. Non-communicable disease (NCD) risk factors among the urban poor in Bangladesh - Preliminary findings of a community based qualitative study . In: Proceeding of the 26th Annual Conference of Japan Society for International Development . Niigata: The 26th Annual Conference Secretariat; 2015 . p. 252 - 5 .